Reports

Germline Mutations in Genes Within the MAPK Pathway Cause Cardio-facio-cutaneous Syndrome

See allHide authors and affiliations

Science  03 Mar 2006:
Vol. 311, Issue 5765, pp. 1287-1290
DOI: 10.1126/science.1124642

Abstract

Cardio-facio-cutaneous (CFC) syndrome is a sporadic developmental disorder involving characteristic craniofacial features, cardiac defects, ectodermal abnormalities, and developmental delay. We demonstrate that heterogeneous de novo missense mutations in three genes within the mitogen-activated protein kinase (MAPK) pathway cause CFC syndrome. The majority of cases (18 out of 23) are caused by mutations in BRAF, a gene frequently mutated in cancer. Of the 11 mutations identified, two result in amino acid substitutions that occur in tumors, but most are unique and suggest previously unknown mechanisms of B-Raf activation. Furthermore, three of five individuals without BRAF mutations had missense mutations in either MEK1 or MEK2, downstream effectors of B-Raf. Our findings highlight the involvement of the MAPK pathway in human development and will provide a molecular diagnosis of CFC syndrome.

There is an emerging group of medical genetic syndromes that are due to activating mutations in genes associated with the Ras pathway, including Noonan syndrome (NS, PTPN11) (1) and Costello syndrome (CS, HRAS) (2, 3). Cardio-facio-cutaneous syndrome [CFC; Online Mendelian Inheritance in Man (OMIM) 115150] has many features that overlap with NS and CS. CFC is a sporadic, complex developmental disorder involving characteristic craniofacial features, cardiac anomalies (most commonly atrial septal defect and pulmonic stenosis), hair and skin abnormalities, postnatal growth deficiency, hypotonia, and developmental delay (4). Because of the similarity between CFC and CS, we screened patients with CFC for mutations in HRAS (3). We found no mutations in this gene, supporting a distinct genetic etiology between CS and CFC syndromes. We therefore expanded our search and sequenced other Ras genes (see Materials and Methods in the supporting online material), as well as genes encoding downstream effectors of Ras: BRAF, CRAF, MEK1, and MEK2. Our cohort consisted of 23 unrelated individuals with the clinical diagnosis of CFC syndrome who did not have a mutation in HRAS or PTPN11 (table S1).

Using bidirectional sequencing of peripheral lymphocyte genomic DNA, we identified heterogeneous missense mutations in BRAF [GenBank accession (NM) 004333] in 18 out of 23 (78% of) individuals having CFC syndrome. Eleven distinct missense mutations clustered in two regions (Fig. 1A). Five individuals had a nucleotide (nt) switch, specifically nt770A→G transition in exon 6, with a predicted missense substitution of arginine for Gln257 (Q257R) (5) in the cysteine-rich domain (CRD) of the conserved region 1 (CR1) (Fig. 1B). The other cluster of mutations was in the protein kinase domain and involved exons 11, 12, 14, and 15. Five patients had heterogeneous missense mutations in exon 12 (table S2). Mutations identified at a lower frequency included missense mutations in the glycine loop encoded by exon 11 (n = 3), the catalytic domain encoded by exon 14 (n = 1), and the DFG motif in the activation segment (exon 15; n = 3). All parents and controls, totaling 40 phenotypically unaffected individuals, had none of these mutations, which supports the hypothesis that the occurrence of CFC is sporadic.

Fig. 1.

Detection of BRAF, MEK1, and MEK2 mutations. (A) Schematic diagram of BRAF showing causative mutations identified in CFC syndrome. BRAF is located on chromosome 7q34 and contains 18 exons with intervening sequences (not drawn to scale). The start and stop codons are indicated. There are three conserved regions. CR1 contains the Ras-binding domain (RBD) and the cysteine-rich domain (CRD), both of which are required for recruitment of B-Raf to the cell membrane. CR2 is the smallest of the conserved regions and CR3 contains the kinase domain. Also depicted are the glycine-rich (G-) loop (exon 11) and the activation segment (exon 15) of the catalytic domain. The 11 missense mutations identified in our cohort of individuals with CFC syndrome are depicted. (B) Lymphocyte DNA electropherograms of a proband and parents are shown identifying a BRAF missense mutation in exon 6 in the proband. Parental DNA samples show normal wild-type sequences. (C) Schematic diagram of genes MEK1 and MEK2 showing causative mutations identified in CFC syndrome. The MEK1 gene is located on chromosome 15q22.31, and MEK2 is on chromosome 19p13.3. Each gene contains 11 exons with intervening sequences (not drawn to scale). Missense mutations identified in three individuals with CFC syndrome are depicted. (D) Lymphocyte DNA electropherograms of the proband and parents are shown identifying the MEK2 missense mutation in exon 2. Parental DNA samples show normal wild-type sequences.

Although the causative mutations were heterogeneous, the distribution of mutations was specific and nonrandom. No frameshift, nonsense, or splice mutations were detected in the cohort of patients; thus, BRAF haploinsufficiency is not a likely causative mechanism of CFC. In 5 out of 23 (22% of) individuals with CFC syndrome, no BRAF mutations were identified. Three of these individuals were found to have missense mutations in MEK1 (NM 002755) and MEK2 (NM 030662) that encode downstream effectors of B-Raf (Fig. 1C). Two individuals had heterogeneous mutations in MEK1: nt158T→C transition (F53S) and a nt389A→G transition (Y130C) in the protein kinase domain (fig. S1). CFC patient number 21 had a MEK2 missense nt170T→G transversion, predicting a F57C substitution (Fig. 1D). Interestingly, F57 in MEK2 (MAPK kinase 2) is the equivalent position to F53 in the closely related MEK1, which suggests that substitutions of this residue may have similar functional consequences in the two family isoforms. The disease-causing mutations in the other two individuals remain to be identified.

The Raf/MEK/ERK cascade is the best understood of the MAPK pathways. (ERK, the extracellular signal–regulated kinase, is a type of MAPK.) In addition to B-Raf, the Raf family includes C-Raf-1 and the X-linked A-Raf. The expression pattern of each isoform is distinct (6), and genetic studies in mice have revealed nonredundant developmental functions (7, 8). Somatic mutations in BRAF occur at high frequency in numerous human cancers (9). One mutation, V600EB-Raf, which confers increased kinase activity, accounts for more than 90% of these mutations. Its presence in benign nevi, as well as primary and metastatic melanoma, suggests that MAP kinase activation is important in melanocytic neoplasia but insufficient for tumorigenesis (10).

In contrast to the mutation spectrum seen in cancer, the BRAF missense mutations identified in individuals having CFC syndrome are more widely distributed (Fig. 1A). Of the 11 different missense amino acid substitutions, only five involve codons that are altered in cancer (table S2), yet only two individuals with CFC syndrome, both of whom have severe phenotypes, have the same substitution that has been reported in cancer (fig. S2). To explore the functional consequences of these mutations, we compared the kinase activity of the CFC B-Raf mutants to that of the wild-type protein (WTB-Raf) and several cancer-derived mutants in transfected human embryonic kidney 293T cells (Fig. 2A; SOM). Four of the CFC B-Raf mutants had increased kinase activity compared with WTB-Raf, and this activity was as high as that of the V600EB-Raf mutant found in cancer. Two CFC B-Raf mutants had lower activity than WTB-Raf and appear to be kinase-impaired (11). Thus, the type of B-Raf mutations found in CFC recapitulates the different types of mutations found in cancer, those with high kinase and kinase-impaired activities. To determine the ability of CFC B-Raf mutants to activate downstream effectors, we measured phosphorylated species of MEK and ERK in transfected cells by Western blotting (Fig. 2B; SOM). Both cancer- and CFC-associated B-Raf mutants with elevated kinase activity (Fig. 2A) induced higher levels of MEK and ERK phosphorylation compared with WTB-Raf, whereas kinase-impaired B-Raf mutants were impaired in their ability to induce phosphorylation of MEK and ERK (Fig. 2B).

Fig. 2.

Functional characterization of B-Raf and MEK mutants identified in CFC. (A) Kinase activities of B-Raf missense CFC mutants (black bars) are compared with those of B-Raf mutations found in cancer (gray bars). Empty vector, wild type B-Raf (WTB-Raf), or the indicated B-Raf point mutants were transfected in 293T cells and B-Raf activity was measured on Flag-immunoprecipitates using a coupled MEK-ERK assay with myelin basic protein (MBP) as the final substrate. Error bars represent the standard deviation of duplicates. (B) MEK and ERK activation by B-Raf mutants. Lysates from 293T cells transfected with empty vector, WTB-Raf, or B-Raf mutants were subjected to SDS–polyacrylamide gel electrophoresis (SDS-PAGE), and MEK and ERK (p44 ERK1 and p42 ERK2) phosphorylation levels were assayed by Western blotting using phospho-specific antibodies. Total ERK antibody staining is shown as a loading control. (C) ERK activation by F53SMEK1 and Y130CMEK1 CFC mutants. ERK phosphorylation induced by the CFC MEK1 mutants were compared with that induced by empty vector, WTMEK1, K97MMEK1, which is a kinase inactive mutant, and S218D/S222DMEK1, which is a constitutively active mutant (13). ERK phosphorylation (p44 ERK1 and p42 ERK2) was assayed by Western blotting using phospho-specific antibodies. Hemagglutinin (HA)–tagged MEK1 is shown as a measure of transfection efficiency, and total ERK is shown as a loading control. (D) ERK activation by the F57CMEK2 CFC mutant. ERK phosphorylation induced by the CFC F57CMEK2 mutant is compared with that induced by empty vector, WTMEK2, K101MMEK2, which is a kinase inactive mutant, and S222D/S226DMEK2, which is a constitutively active mutant (13). ERK phosphorylation (p44 ERK1 and p42 ERK2) was assayed by Western blotting using phospho-specific antibodies. Myc-tagged MEK2 is shown as a measure of transfection efficiency, and total ERK is shown as a loading control.

Missense mutations in MEK1 and MEK2, which encode the only known effectors of B-Raf, also cause CFC syndrome. MEK1 and MEK2 are dual-specificity kinases that both activate ERK1 and ERK2 but appear to play nonredundant roles. Genetic evidence from mouse models indicates that MEK1 is essential for embryonic development (12), whereas MEK2 is dispensable (13). Although activation of MEK is necessary for mammalian cell transformation through the MAPK cascade (14) and constitutively active MEK mutants promote transformation (15), mutations in MEK1 and MEK2 have thus far not been reported in human cancer (9, 16). Three individuals in the CFC cohort (13%) had de novo missense mutations, with two mutations occurring in equivalent positions within exon 2 of MEK1 and MEK2 (Fig. 1C). To explore the functional consequences of these substitutions, we assayed ERK phosphorylation in transfected cells by Western blotting (Fig. 2, C and D). All CFC MEK mutants, F53SMEK1, Y130CMEK1, and F57CMEK2 were more active than wild-type MEK in stimulating ERK phosphorylation, but they were not as active as the constitutively active MEK mutants. Although our current CFC cohort with MEK1/2 mutations is few in number, the phenotypic features of individuals are concordant with those observed in mouse models (fig. S1). Transgenic mice expressing activated MEK1 have enhanced MEK1-ERK1/2 signaling and exhibit compensated cardiac hypertrophy (17), hyperkeratosis and epidermal hyperproliferation (18, 19), and cataract formation (20).

CFC syndrome, a developmental disorder that is phenotypically similar to NS and CS, is unique in that it is caused by missense mutations in one of three different signaling components of the Ras/MAPK pathway (Fig. 3). Interestingly, unlike NS or CS, CFC syndrome has not been considered a cancer-predisposing syndrome, because individuals do not develop malignancies. Our findings highlight the involvement of the MAPK pathway in human development. Individuals with the suspected clinical diagnosis of CFC syndrome can now be diagnosed on a molecular basis. Because the MAPK pathway has been studied intensively in the context of cancer, therapeutics that specifically target this pathway are in development. Inhibitors of Raf and MEK are being evaluated in clinical trials and appear to be well tolerated (21). In addition, a recent report indicates that cells with activated B-Raf have enhanced, selective sensitivity to MEK inhibitors (22). Because CFC has an evolving phenotype, systemic therapies that reduce MAPK activity may merit investigation in this population of patients.

Fig. 3.

Ras/Raf/MEK/ERK signal transduction pathway and associated genetic syndromes.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1124642/DC1

Materials and Methods

Figs. S1 and S2

Tables S1 and S2

References

References and Notes

View Abstract

Navigate This Article